Clinical StudyTranspancreatic Precut Sphincterotomy for Biliary Access:The Relation of Sphincterotomy Size to Immediate SuccessRate of Biliary Cannulation

Lien-Fu Lin

Division of Gastroenterology, Department of Internal Medicine, Tungs’ Taichung Metroharbor Hospital,Section 8, No. 699 Taiwan Boulevard, Wuqi District, Taichung 435, Taiwan

Correspondence should be addressed to Lien-Fu Lin; aucardla@yahoo.com.tw

Received 1 December 2013; Revised 22 January 2014; Accepted 6 February 2014; Published 10 March 2014

Academic Editor: Everson Artifon

Copyright © 2014 Lien-Fu Lin. This is an open access article distributed under the Creative Commons Attribution License, whichpermits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background. Transpancreatic precut sphincterotomy (TPS) is an option for difficult commonbile duct (CBD) access, and the reportsare few, with immediate success rate varying from 60 to 96%.The description of relation between the size of TPS and the immediatesuccess rate of CBD cannulation was not found in the literature. The Aim of the Study. To evaluate the relation of large TPS toimmediate success rate of CBD cannulation. Methods. A retrospective analysis was performed in prospectively collected data of20 patients. TPS was performed with traction papillotome in the main pancreatic duct (MPD) directing towards 11 o’clock. Needleknife (NK) was used to enlarge TPS in five patients, and the other 15 cases had large TPS from the beginning of sphincterotomy.Prophylactic pancreatic stent was inserted in 18 cases, with diclofenac given in 12 cases. Results. The immediate success rate of CBDcannulation was 90% and with an eventual success rate of 100%. The failure in one immediate CBD cannulation with large TPSwas due to atypical location of CBD orifice, and the other failed immediate CBD cannulation was due to inadequate size of TPS.Complications included 3 cases of post-TPS bleeding and 3 cases of mild pancreatitis. Conclusion. TPS is an effective procedure inpatients with difficult biliary access and can have high immediate success rate with large TPS.

1. Introduction

The initial success rate of selective CBD cannulation wasreported to be 70% after a standard cannulation using guidewire and papillotome with the criteria of 5 attempts ofcannulation [1]. Using double-guide wire (DGW) technique,the success rate varies from 47 to 73% with the pancreatitisrate of 14–38% [1–3]. Transpancreatic precut sphinctero-tomy (TPS) can be an alternative technique because theaccess is already in the MPD. The reports of TPS arefew when compared to needle knife precut sphincterotomy,and detailed description of the size of TPS in relation toimmediate successful CBD cannulation was not emphasizedin the literature. The immediate success rate of TPS forbiliary access varies from 60 to 96% [4, 5]. Diclofenacrectally administered and prophylactic pancreatic stent (PS)

placement can reduce the incidence of postendoscopic ret-rograde cholangiopancreatography pancreatitis (PEP) [6].We evaluate the primary outcome of immediate successrate of CBD cannulation with large TPS and secondaryoutcome of the complications with this new technique in ourhospital.

2. Materials and Methods

Between January 2012 and October 2013, a retrospectiveanalysis of prospectively collected data of 20 cases of TPSwith inaccessible bile duct was conducted.There were 13maleand 7 female patients with a mean age of 54 ± 16 (25–82)years. The procedure was performed by a single operatorwho has experience on ERCP and endoscopicsphincterotomy

Hindawi Publishing CorporationDiagnostic and erapeutic EndoscopyVolume 2014, Article ID 864082, 7 pageshttp://dx.doi.org/10.1155/2014/864082

2 Diagnostic andTherapeutic Endoscopy

(a) (b)

(c) (d)

Figure 1: (a) Deviated papilla (arrow), (b) transpancreatic precut sphincterotomy in 11 o’clock direction (arrow), (c) CBD orifice (arrow)located left to the pancreatic duct occupied by sphincterotome, and (d) CBD stone extracted (arrow).

(EST) for more than 1000 cases. A total of 112 cases ofendoscopic retrograde cholangiopancreatography (ERCP)were performed during that period when TPS was startedas a precut technique. The inclusion criteria were (1) failedselective deep CBD cannulation after 20 minutes and (2)successful access in the MPD. The exclusion criteria were(1) post-Billroth II gastrectomy cases and (2) failed accessto the MPD. After access to the MPD, TPS was performedwith the traction sphincterotome directing towards 11 o’clock,and CBD orifice is usually located left to the pancreatic ductafter TPS (Figures 1 and 2). Needle knife was required toextend the TPS over the pancreatic stent (Figure 3) to getbiliary access due to inadequate sphincterotomy in 5 cases.Large TPS incision up to superior margin of the papilla wasperformed in the other 15 cases (Figure 4). Pancreatic stent(PS) was placed either before or after the complete biliaryprocedure. The PS was removed after 4-5 days when therewere no post-TPS bleeding and no spontaneous passage ofthe pancreatic stent. Diclofenac was given in patients who

had balloon dilatation of the papilla or contrast injection intomain pancreatic duct. Written consents were obtained fromall of the patients. After TPS, serum amylase was tested on thenext day and repeated on the 3rd day if the amylase level waselevated 3 times above the normal limit.

The instruments and current setting used were (1) duo-denoscope (TJF 240, Olympus Optical Co., Ltd., Tokyo,Japan), (2) sphincterotome (Olympus KD-431Q-0720), (3)Boston ScientificMicrovasive Jagwire, (4)WilsonCookZim-mon pancreatic stent 5Fr-4 cm, (5) Wilson Cook HuibregtseHPC-3 needle knife, (6) electrosurgical generator (OlympusPSD 60), and (7) current setting (Endocut setting output of120Watts and effect 2).

The definitions were as follows: (1) inaccessible bile ductwas defined as failure after 20 minutes of CBD cannulation[7, 8], (2) complications of pancreatitis and bleeding weredefined according to the criteria used by Cotton et al. [9], and(3) large incision sphincterotomy was defined as incision upto superior margin of the bulging papilla [4].

Diagnostic andTherapeutic Endoscopy 3

(a) (b)

(c) (d)

Figure 2: (a) Long papilla (arrow), (b) transpancreatic precut sphincterotomy in 11 o’clock direction (arrow), (c) CBD cannulation bysphincterotome (arrow), and (d) postbiliary and pancreatic stenting (white and blue arrows).

3. Results

The results were shown in Table 1. The etiologies were 14cases of common bile duct stone, 3 pancreatic head cancers,1 bile leak, 1 biliary dilatation, and 1 chronic pancreatitis.The reasons for difficult selective CBD cannulation wereperiampullary diverticulum, long, and/or deviated papilla.Three cases had undergone trial of CBD cannulation withDGW technique but failed to get biliary access. Five casesneeded additional needle knife sphincterotome to enlarge theTPS incision. All cases of large TPS incision had successfulinitial CBD cannulation except one owing to atypical locationof CBD orifice (Figure 5).The immediate success rate of CBDcannulation after TPS was 90% and the eventual success ratewas 100%. The other failed immediate CBD cannulation wasdue to inadequate size of TPS.

One massive delayed bleeding required surgery withuneventful recovery after failed endoscopic hemostasis andtransarterial embolization. This severe bleeding case hadundergone large TPS incision. One mild immediate bleeding

was treated with endoscopic local epinephrine injection andclipping. The other mild repeated bleeding was managedendoscopically as well.Three cases ofmild pancreatitis recov-ered within 2 days, though they had received prophylacticpancreatic stenting, and 2 cases had diclofenac given rectally.There was no perforation complication.

4. Discussion

Techniques for difficult deep CBD cannulation include guidewire technique, DGW technique, and precut access. The suc-cess rate of CBD cannulation is not high in DGW techniqueranging from 47 to 73% but the pancreatitis rate can be ashigh as 38% [1, 2]. Our unpublished success rate with DGWtechnique was 60% with a mild pancreatitis of 6%. Precuttechnique can have needle knife precut (free hand cutting atthe papillary orifice or cutting the papilla over a pancreaticstent), needle knife fistulotomy above papillary orifice, andtranspancreatic precut sphincterotomy (TPS) using tractionpapillotome [10].

4 Diagnostic andTherapeutic Endoscopy

(a) (b)

(c) (d)

Figure 3: (a) Papilla (arrow), (b) transpancreatic precut sphincterotomy in 11 o’clock direction (arrow) with inadequate incision, (c) needleknife used to enlarge the incision (arrow), and (d) CBD cannulation (arrow).

TPS was first described by Goff et al. [11], and very fewreports about comparison between TPS with needle knifesphincterotomy can be found [12]. TPS was reported tohave a higher success rate and less complication comparedwith standard needle knife precut [13] and also a safe,effective procedure in patients with difficult bile duct accesswhere classical sphincterotomy or needle knife proceduresfail [14]. But on the other hand, Wang et al. conducted amulticenter prospective trial of TPS with needle knife precutsphincterotomy and concluded that there is no significantdifference in success rate and complications [15].

The immediate success rate of TPS varies from 60 to 96%[4, 5]. No specific emphasis on or description about the sizeof TPS with the immediate success rate could be found in theliterature. Although Goff reported a size of 5–7mm of TPSachieving a success rate of 96%, but an initial success rate ofonly 60% was described by Akashi et al. with medium andlarge TPS [4]. Kapetanos et al. described an initial successrate of 75% without mentioning the exact length of TPS [16].High initial success rate of 97% with large TPS was achieved

by Halttunen et al. [17], and Kahaleh et al. also reported 85%success rate with large TPS [18]. Out of 20 cases of TPS inour hospital, 14 cases of large TPS incision at first attemptresulted in immediate successful CBD cannulation and fasterprocedure because no additional needle knife enlargementwas needed. One large TPS that failed to get the biliaryaccess was attributed to the atypical location of CBD orifice.Therefore, the key to a higher successful immediate biliaryaccess is to incise a large TPS at the first attempt. The reasonof requiring additional needle knife for extension of TPS in 5cases was due to inadequate incision of TPS. Our immediatesuccess rate of CBD cannulation was 90%, and eventualsuccess rate was 100%.

The incidence of post-TPS bleeding varies from 3.7 to5% [7, 16], our post-TPS bleeding was 15% which was a littlehigher and may be due to a small number of cases. Massivepost-TPS bleeding was not reported in the literature. Severepostsphincterotomy bleeding was presumed to result fromincision through an aberrant retroduodenal artery [19, 20],and the length of incision is not an important predictor of

Diagnostic andTherapeutic Endoscopy 5

(a) (b)

(c) (d)

Figure 4: (a) Transpancreatic precut sphincterotomy in 11 o’clock direction (arrow), (b) large incision sphincterotomy (arrow), (c) successfulCBD cannulation (arrow), and (d) postbiliary and pancreatic stenting (white and blue arrows).

(a) (b)

Figure 5: (a) Posttranspancreatic precut sphincterotomy with pancreatic stenting (arrow) and (b) CBD orifice located on the right side ofpancreatic duct (arrow).

6 Diagnostic andTherapeutic Endoscopy

Table 1

No. Age Sex Diagnosis Papilla DGW NKS Success Bleeding P stent Voren Panreatitis1 49 M Pan cancer Long — — Yes 1 mild Yes — —2 47 F CBDS Long — — Yes No Yes 100 —3 44 M Bile leak Long — — Yes 1 massive# Yes — —4 45 F CBDS Long, deviated Yes — Yes No Yes — —5 45 M CBDS Long, deviated — Yes Yes 1 mild Yes — Mild6 32 F CBDS Deviated Yes Yes Yes No Yes — —7 77 M CBDS Long — Yes Yes∗ No Yes 75 —8 39 M CCP Long — — Yes No Yes 100 —9 25 M CBDS Deviated — Yes Yes No Yes 100 —10 72 M CBDS Long Yes — Yes No Yes 75 —11 56 F CBDS Long, deviated — — Yes No Yes — —12 56 F CBDS Long, deviated — — Yes No Yes 100 Mild13 58 M CBDS Deviated — — Yes No Yes 75 —14 56 M Pan ccncer Deviated — — Yes No 0 — —15 82 M CBDS PAD — Yes Yes No Yes 50 —16 70 F BD (pap bx) Long, deviated — — Yes No Yes 75 Mild17 73 M Pan cancer Deviated — — Yes No 0 — —18 70 M CBDS Long, deviated — — Yes No Yes 100 —19 64 M CBDS Deviated — — Yes No Yes 100 —20 29 F CBDS PAD — — Yes∗ No Yes 100 —DWG: double guide wire technique, NK: needle knife, P: pancreatic, Diclo: diclofenac, CBDS: common bile duct stone, BD: bile duct dilatation, Pap: papillary,PAD: periampullary diverticulum, ∗Second attempt of CBD cannulation, #massive delayed TPS bleeding ended with surgery after failed endoscopic hemostasisand transarterial emboiization.Long papilla: >2 cm and with difficult selective CBD cannulation.Deviated papilla: difficult to adjust the papillary orifice in an en-face position.Lower dose of diclofenac was given in older patients with mildly elevated renal function.

bleeding [21]. The cause of massive post-TPS (initial largeTPS) bleeding in our series was attributed to have inciseda larger branch of gastroduodenal artery after reviewing theangiography. More cases of TPS are needed for evaluation forpost-TPS bleeding.

The incidence of pancreatitis without prophylactic stent-ing after TPS ranged from 5.5 to 21% [4, 7, 13] and reducedto 3.5% if prophylactic pancreatic stenting was performed[5]. We had 3 cases (15%) of mild pancreatitis, but all ofthem recovered in 2 days. All 3 cases had prophylacticpancreatic stenting done, and two had received diclofenacadministration. The cause of pancreatitis in our series couldbe due to contrast injection intoMPDor deepGWplacementin the MPD. Severe post-TPS pancreatitis can occur withoutpancreatic stenting [3, 13]; therefore, pancreatic stenting isnecessary in TPS procedure [6].

The advantages of large TPS are (1) definite completeunroofing of the papilla that makes CBD cannulation easierenabling a higher immediate success rate of biliary access and(2) probably low incidence of perforation which has beenreported in two studies [5, 18]. The other credits of TPS areless technically demanding and easier to control the depth ofcutting [13].The limitations in this study are (1) small numberof cases and (2) nonrandomized study for median or largeTSP incision for comparison of immediate CBD cannulationsuccess rate.

5. Conclusion

TPS is an effective procedure in patients with difficult biliaryaccess and can have high immediate success rate with largesized TPS.

Consent

Written informed consent was obtained from patients for thepublication of this paper.

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper.

References

[1] A. Herreros de Tejada, J. L. Calleja, G. Dıaz et al., “Double-guidewire technique for difficult bile duct cannulation: amulticenter randomized, controlled trial,” GastrointestinalEndoscopy, vol. 70, no. 4, pp. 700–709, 2009.

[2] K. Ito, N. Fujita, Y. Noda et al., “Pancreatic guidewire placementfor achieving selective biliary cannulation during endoscopicretrograde cholangio-pancreatography,” World Journal of Gas-troenterology, vol. 14, no. 36, pp. 5595–5600, 2008.

Diagnostic andTherapeutic Endoscopy 7

[3] Y. W. Yoo, S. W. Cha, W. C. Lee et al., “Double guidewiretechnique vs transpancreatic precut sphincterotomy in difficultbiliary cannulation,” World Journal of Gastroenterology, vol. 19,no. 1, pp. 108–114, 2013.

[4] R. Akashi, T. Kiyozumi, K. Jinnouchi, M. Yoshida, Y. Adachi,and K. Sagara, “Pancreatic sphincter precutting to gain selectiveaccess to the common bile duct: a series of 172 patients,”Endoscopy, vol. 36, no. 5, pp. 405–410, 2004.

[5] J. S. Goff, “Long-term experience with the transpancreaticsphincter pre-cut approach to biliary sphincterotomy,” Gas-trointestinal Endoscopy, vol. 50, no. 5, pp. 642–645, 1999.

[6] J.-M. Dumonceau, A. Andriulli, J. Deviere et al., “EuropeanSociety of Gastrointestinal Endoscopy (ESGE) Guideline: pro-phylaxis of post-ERCP pancreatitis,” Endoscopy, vol. 42, no. 6,pp. 503–515, 2010.

[7] H. Miyatani and Y. Yoshida, “Endoscopic needle knife precutpapillotomy for inaccessible bile duct following failed pancre-atic duct access,” Clinical Medicine, vol. 2009, no. 2, pp. 1–5,2009.

[8] K. W. Yoon, C. H. Park, S. Y. Park et al., “Guidewire cannu-lation increases the success rate of needle-knife fistulotomyfor difficult bile duct access,” Journal of Gastroenterology andHepatology, vol. 25, no. 1, pp. 14–18, 2010.

[9] P. B. Cotton, G. Lehman, J. Vennes et al., “Endoscopic sphinc-terotomy complications and their management: an attempt atconsensus,” Gastrointestinal Endoscopy, vol. 37, no. 3, pp. 383–393, 1991.

[10] M. L. Freeman and N. M. Guda, “ERCP cannulation: a reviewof reported techniques,” Gastrointestinal Endoscopy, vol. 61, no.1, pp. 112–125, 2005.

[11] J. S. Goff, “Common bile duct pre-cut sphincterotomy: trans-pancreatic sphincter approach,”Gastrointestinal Endoscopy, vol.41, no. 5, pp. 502–505, 1995.

[12] T. DaVee, J. A. Garcia, and T. H. Baron, “Precut sphincterotomyfor selective biliary duct cannulation during endoscopic retro-grade cholangiopancreatography,” Annals of Gastroenterology,vol. 25, pp. 1–12, 2012.

[13] M. F. Catalano, J. D. Linder, and J. E. Geenen, “Endoscopictranspancreatic papillary septotomy for inaccessible obstructedbile ducts: comparison with standard pre-cut papillotomy,”Gastrointestinal Endoscopy, vol. 60, no. 4, pp. 557–561, 2004.

[14] A. Weber, T. Roesch, S. Pointner et al., “Transpancreatic precutsphincterotomy for cannulation of inaccessible common bileduct: a safe and successful technique,” Pancreas, vol. 36, no. 2,pp. 187–191, 2008.

[15] P.Wang,W. Zhang, F. Liu et al., “Success and complication ratesof two precut techniques, transpancreatic sphincterotomy andneedle-knife sphincterotomy for bile duct cannulation,” Journalof Gastrointestinal Surgery, vol. 14, no. 4, pp. 697–704, 2010.

[16] D. Kapetanos, G. Kokozidis, D. Christodoulou et al., “Caseseries of transpancreatic septotomy as precutting technique fordifficult bile duct cannulation,” Endoscopy, vol. 39, no. 9, pp.802–806, 2007.

[17] J. Halttunen, I. Keranen, M. Udd, and L. Kylanpaa, “Pancreaticsphincterotomy versus needle knife precut in difficult biliarycannulation,” Surgical Endoscopy andOther Interventional Tech-niques, vol. 23, no. 4, pp. 745–749, 2009.

[18] M. Kahaleh, J. Tokar, T. Mullick, S. J. Bickston, and P. Yeaton,“Prospective evaluation of pancreatic sphincterotomy as a pre-cut technique for biliary cannulation,” Clinical Gastroenterologyand Hepatology, vol. 2, no. 11, pp. 971–977, 2004.

[19] J. W. C. Leung, F. K. L. Chan, J. J. Y. Sung, and S. C. S.Chung, “Endoscopic sphincterotomy-induced hemorrhage: astudy of risk factors and the role of epinephrine injection,”Gastrointestinal Endoscopy, vol. 42, no. 6, pp. 550–554, 1995.

[20] J. E. Geenen and J. A. LoGuidice, “Endoscopic sphincterotomyfor diseases of the biliary tree,” Advances in surgery, vol. 14, pp.31–51, 1980.

[21] M. L. Freeman, D. B. Nelson, S. Sherman et al., “Complicationsof endoscopic biliary sphincterotomy,”TheNewEngland Journalof Medicine, vol. 335, no. 13, pp. 909–918, 1996.